Sasha Issenberg: LibraryThing Author Interview
Sasha Issenberg is the "Victory Lab" columnist for Slate and the Washington correspondent for Monocle. He covered the 2008 election for the Boston Globe, and his work has been appeared in many publications. His second book is The Victory Lab: The Secret Science of Winning Campaigns, published this month by Crown.
For those who might not have had the chance to read The Victory Lab yet, give us the nutshell version: What's the book about?
It's the story of a scientific revolution that—almost entirely outside of the view of the public and political media—has upended the way campaigns are run. The book jacket has a quote from Politico that calls it "Moneyball for politics," and I think that's not mere marketing bluster. The Victory Lab follows a very similar narrative of a culture clash transforming a traditionally minded industry. In my story, like Michael Lewis's, a generation of geeks using fresh data and analytical techniques challenges the tyranny of an old guard reliant on instinct and lore—the only difference is that the old guard are campaign consultants, not scouts, and the prize isn't the World Series but the White House.
You discuss various tactics used to get more people to the polls in a given election, and you note that one particular method has proven to be of great effectiveness but tends not to get used at all. Why do you think candidates and parties have shied away from publicly sharing voting statistics and patterns, even though they know it does get people to the polls?
The most effective way of turning a non-voter into a voter—several times more effective than any other technique that's ever been measured—is to send a citizen a copy of her vote history, the record of elections in which she cast a ballot, and her neighbors' vote histories. Then tell her that everyone will get an updated set after the coming election. Behavioral psychologists call this "social pressure," the idea that people adjust their behavior to conform with what they think are others' expectations. In 2006, a few researchers ran a field experiment and found that sending such mail had a massive impact on turnout—but they also got death threats from people who accused them of blackmail.
This is an instance where the knowledge was shared, and, in fact, the study was published in a political science journal. (It's proven to be one of the discipline's ballsier experiments to date.) But political campaigns and parties have been wary of using it for fear of being labeled bullies by voters. Over the following years, political operatives and academics found ways to soften the language, while still exerting subtle social pressure and impacting voter turnout—and these results are likely to hit millions of mailboxes before Election Day.
Are you seeing major differences in campaign tactics and strategies in the wake of the Citizens United decision, or are campaign tactics separate from the broader concerns about the amount of money being thrown around?
Much of the money spent by super-PACs seems to be on fairly conventional TV ad buys. One of the themes that came out of my reporting on the last decade in politics is how much this new science has awakened an appreciation within campaigns for volunteer-based, person-to-person communication. This is not just because charismatic candidates like George W. Bush and Barack Obama have motivated their parties' activists to give their time, but because parties have recognized the value of data collected by volunteers, and through experiments the impact that their activity can have on voter behavior. And all of this goes through parties or candidate campaigns. After all, people may be writing checks to a super-PAC, but no one volunteers for one.
It's the same sensibility, and the spine of data and analytics for targeting voters remains in place. One thing that is new, and that I've written about for Slate since the book went to press, is the way the campaign is integrating field experiments to test messages. These are basically the campaign version of a pharmaceutical trial. Instead of randomly assigning drugs and placebos, the campaign distributes political messages. To do this, Obama has contracted the Analyst Institute, a secret society of researchers from Democratic Party politics and academia whose success bringing experimental methods to campaigns is a major storyline in The Victory Lab.
Because Obama's headquarters has thousands of individual data points on the voters in its sample, analysts there can make a very granular determination about what types of voters actually move in response to which messages. This groundbreaking technique is a radical departure from relying solely on traditional polls and focus groups, and the fact that a presidential campaign relies on it demonstrates how in many respects the revolutionaries I write about in the book are now running the show.
We've seen some interesting studies being done on campaign tactics this year, particularly ProPublica's examination of campaign emails. Is this the first time these sorts of tactics are being scrutinized in real time, and do you think it will have any effect on how campaigns use these marketing techniques?
I think campaigns are generally aware that while the Internet gives them many more channels for highly individualized communication, it also makes it a lot easier for voters and the media to pick up on it. It's hard to keep things narrowly targeted online. It's a lot easier to instantly forward around a provocative email or tweet a screenshot of a controversial Web ad than it was to let the world know about the direct mail you've received.
But the limit of projects like ProPublica is we still don't know how widely these emails were sent, to what segments of the campaign's database, and if they were intended to test certain fund-raising appeals or represented a campaign rolling out one that has already tested well. Those are the questions that can give real insight into strategy, and both campaigns are pretty confident that the media will never be able to answer those questions and decode their activities.
What should we as citizens know and understand about how candidates and political parties are gathering information about us and how they're using it?
Campaigns have access to thousands of pieces of individual data on every adult in the United States. Some comes from publicly available documents like your voter-registration record, others are derived from what the Census says about your block or neighborhood, and a lot of it is acquired from commercial data vendors—the same people who supply credit-card companies and mail-order catalogues with information to direct their marketing efforts. Some of the most valuable data, however, comes from political intelligence that citizens readily give up: what you tell a canvasser or phone-bank volunteer about the candidate you prefer or the issue that matters most to you will now likely live on for decades in the parties' databases.
Analysts build statistical models off all of this data, relying on algorithms to detect otherwise imperceptible relationships between election-year opinions and individual attributes. The result is a final product that is something like your individual credit score, but for politics: a campaign's prediction of how they think you are likely to vote, how probable you are to cast a ballot, and how much you care about a particular issue.
Now that you've researched this topic so thoroughly, do you think there are a few simple steps that might make voters better informed, and to encourage more people to participate in elections and take a greater interest in the choices they make politically?
I think we naturally tend to pose this question. But instead of asking why people don't get involved in politics—typically wondering why citizens choose not to vote—I'm a little more curious as to why so many do. This question has puzzled economically minded political scientists for a while, because in many cases voting seems irrational: It costs a lot of the voter and offers little in the way of material reward.
The behavioral-science revolution in politics suggests that we have been underestimating how much citizens' motivation for being involved isn't about an election's outcomes but about what it means to us to be involved. That's what was most radical about the social-pressure experiment in Michigan: It dramatically changed citizens' desire to vote without giving them any new information about the candidates or the issues or the stakes. All those postcards did was give citizens information about themselves and their neighbors.
What's your library like? What kinds of books would we find on your shelves?
I read almost exclusively nonfiction. A lot of what is on my shelves can be classified as history, both popular and academic. I'm particularly drawn to cities and urbanism, so a lot of books I own satisfy that curiosity in some form or another.
Which books have you read and enjoyed recently?
Like anyone fascinated by politics, I'm in awe of Robert Caro and liked his latest book on LBJ, The Passage of Power, as much as I expected I would. I spent a weekend in Berlin this summer, so I brought along Erik Larson's In the Garden of Beasts. Both Caro and Larson are excellent at exhuming sensory detail from historical primary sources, one area where I'd very much like to improve as a writer.
I also recently stumbled upon a series of books by a guy named John R. Frears, who appears to have been a British academic who, during the 1970s, specialized in contemporary accounts of French politics. It's a period that interests me greatly, and reading scholarly work in French is still a slog for me, so I really appreciate how graceful and witty Frears' prose is. Added bonus: His France in the Giscard Presidency has a great David Levine presidential caricature on its cover.
Do you have a sense yet of your next book project?
I've already started on it. It will be called The Engagement, a political, legal, and social history of the gay-marriage debate that has moved so quickly over the last quarter-century. It's been a massive shift in American public opinion and has occurred entirely within my lifetime. My goal is to tell a story that explains how and why this happened.
—interview by Jeremy Dibbell
Books by Sasha Issenberg
Recent author interviews
Jeff Greenfield (2013-11-13)
Patrick Ness (2013-10-16)
Tom Standage (2013-10-16)
Samantha Shannon (2013-08-20)
Paul Collins (2013-07-24)
Travis McDade (2013-07-24)
Colum McCann (2013-05-22)
Jennifer McVeigh (2013-05-22)
Julie Wu (2013-05-22)
Marie Brennan (2013-04-26)
Tatiana Holway (2013-04-26)
Elizabeth Strout (2013-03-21)
Kim Ghattas (2013-03-21)
Matthew M. Aid (2013-03-21)
Christine Sneed (2013-02-07)
Robin Sloan (2013-02-07)
Douglas Hunter (2012-12-19)
Simon Garfield (2012-12-19)
Christopher Bonanos (2012-11-20)
Jon Meacham (2012-11-20)
About author interviews
Each month we feature a few exclusive interviews with authors in our "State of the Thing" newsletter. Know an author who might want to be interviewed? Find out more.